Atopic dermatitis and the clinical effect of house dust mite avoidance

Abstract

I have read with great interest the report by Oosting et al.1Oosting AJ De Bruin-Weller MS Terreehorst I Tempels-Pavlica Z Aalberse RC De Monchy JG et al.Effect of mattress encasings on atopic dermatitis outcome measures in a double-blind, placebo-controlled study: The Dutch mite avoidance study.J Allergy Clin Immunol. 2002; 110: 500-506Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar Theirs was a very thoroughly performed prospective investigation of the clinical effect of house dust avoidance strategies on the course of atopic dermatitis (AD). However, I would like to make 5 remarks: 1.The authors performed a subgroup analysis by dividing the patients into 4 subgroups with low and high Der p1 exposures and low and high AD scores. The authors did not observe significant differences in clinical effect among these 4 groups. Surprisingly, there is no subgroup analysis regarding the outcome of the atopy patch test (APT). It has been proposed that APTs indicate the clinical relevance of house dust mite exposure in AD2; this subgroup analysis would therefore have been interesting. Instead, the authors performed an analysis using the APT as secondary outcome measure. This analysis seems to be less useful: A nickel contact dermatitis is expected to improve when the allergen is avoided, but the patch test reaction is not expected to disappear within 1 year.2.The authors state that one of their inclusion criterion was a specific IgE (anti–Der p1) value greater than 0.7 kU/L. However, in Table V there are patients (ranges, 0.2-90 and 0.3-90 kU/L) with values below this threshold.3.The authors state that our previous study on the same topic3Gutgesell C Heise S Seubert S Seubert A Domhof S Brunner E et al.Double-blind placebo-controlled house dust mite control measures in adult patients with atopic dermatitis.Br J Dermatol. 2001; 145: 70-74Crossref PubMed Scopus (105) Google Scholar had “low power” inasmuch as only 10 patients in 2 treatment groups were described. This is not true; in our study, 20 patients (not 10 patients) were investigated. Moreover, the study was powerful enough to show a highly significant effect on reduction of allergen levels, though this was not accompanied by a clinical effect.4.The authors' subgroup analysis (as described above) yielded groups of statistically less than 10 patients (35:4 = 8.75; 38:4 = 9.5). Hence their own analysis, according to their definition, would be of “low power.”5.In conclusion, AD is a heterogeneous disease with (as mentioned above) a number of different subgroups (eg, adult/children, extrinsic/intrinsic type, house dust mite monosensitization/polysensitization, negative/positive atopy patch test, and low/high antigen exposure).4Gutgesell C Heise S Seubert S Seubert A Domhof S Brunner E et al.Atopic dermatitis, house-dust mite, and the placebo effect.Allergy. 2001; 56: 1226-1227Crossref PubMed Scopus (1) Google Scholar It is desirable that in the future studies enrolling a statistically reasonable number of patients be performed in order to stratify for all of these different factors.